Wireless Power for Implasnts

CHARACTERIZATION OF FLEXIBLE ANTENNAS

FOR WIRELESSLY POWERING

MEDICAL IMPLANTS

by

SAFWAN AHMED MUNEER

Presented to the Faculty of the Graduate School of

The University of Texas at Arlington in Partial Fulfillment

of the Requirements

for the Degree of

MASTER OF SCIENCE IN ELECTRICAL ENGINEERING

THE UNIVERSITY OF TEXAS AT ARLINGTON

August 2012

Copyright © by Safwan Ahmed Muneer 2012

All Rights Reserved

iii

ACKNOWLEDGEMENTS

It is with great pleasure that I am writing this thesis work, because it marks the end

of my curriculum requirements for getting my Master’s degree in Electrical Engineering. I am

grateful to Allah for his blessings and my family for their infinite support in pursuing my dream.

Especially the sacrifice made by my mother cannot be forgotten. I am thankful to my late father

whose dream was to get his children the highest quality of education; sadly he is not there to

see his dream achieve fruition.

Without the help and support of Dr. JC Chiao, this work would have been incomplete.

I am grateful to him for financially supporting me and for assisting me in every part of my thesis

work. Dr. Hung Cao played an instrumental role in my thesis work; I would like to take this

opportunity in thanking him. I am grateful to Dr. Sanchali Deb, Uday Shankar Tata, and Mathew

Lee Oseng for helping me be a better individual and for giving me suggestions time and again. I

would like to thank Dr. Mingyu Lu for providing me the technical platform I needed to start my

thesis work. Lastly I would like to thank iMems group whose help and support made my life

easy.

I would like to end my thank you list by extending a special thanks to my thesis

committee members – Dr. W. Alan Davis and Dr. William E. Dillon for reading my thesis and

being in the defense committee.

July 12, 2012

iv

ABSTRACT

CHARACTERIZATION OF FLEXIBLE ANTENNAS

FOR WIRELESSLY POWERING

MEDICAL IMPLANTS

Safwan Ahmed Muneer, M.S.

The University of Texas at Arlington, 2012

Supervising Professor: JC Chiao

To minimize the trouble experienced by the patient in undergoing invasive surgeries to

replace non rechargeable batteries that are used to power body implants inside, medical

experts and engineers are emphasizing on making use of wireless power. Since body implant

powering from external wireless power requires a source that can be easily worn outside the

body, a special type of integrated antenna was implemented on a flexible substrate that

conforms to the shape of the stomach curvature, thus adding increased patient comfort. The

antenna bend is concave which adds to increased focus and power transfer efficiency to power

the body implant. Since the dielectric constant of body cells increases with the increase in the

frequency of operation of the system, the frequency used is 1.3 MHz which is in the ISM band.

Since research is underway to develop a whole new kind of body implants such as retinal

implants, intracranial brain computer interfaces (BCI) etc., the demand for efficient powering of

v

body implants from the outside using a compact source has increased to a great extent.

In this research work, a square planar spiral antenna was fabricated using copper tape and

mask on a flexible substrate – kapton (polyimide) to power the tag antenna (body implant). A

Class – E power amplifier was used to amplify the signal before transmitting it through the

reader antenna to the tag antenna which in practical implementation would be attached to the

body implant. A non - planar antenna made using 24 AWG magnet wire on a Styrofoam

substrate was used to receive the power. Both the reader antenna and tag antenna were

resonated using appropriate capacitors to achieve maximum power transfer. In order to

understand the power transfer pattern in a human body through the stomach, a stable setup

and curvatures of different size were used. Since the reader antenna will be worn in the form of

a belt, it was bent at different radii of curvature. The tag antenna or the reader antenna can

move sideways or radially, experiments were executed to incorporate all these effects and the

results for the power transfer efficiency were recorded. Accidently, it was found that the power

transfer efficiency increased when the reader antenna was bent rather than straight. The

feasibility of the efficiency pattern that is obtained by displacing the reader and tag antennas is

confirmed by fabricating antennas with different designs and obtaining the repeatable pattern.

This pattern can be used as a reference pattern to optimize the parameters of the reader

antenna for optimum power transfer efficiency. Ansoft HFSS, simulation software that makes

use of finite element method (FEM) to perform 3D electromagnetic simulations was used to

simulate the inductance of the reader antenna. The reader antenna 3D model made using the

software was bent to emulate the reader antenna bend along the stomach curvature, using

AutoCAD software and Bonzai 3d software and the model was imported into HFSS and the

simulations were verified with the experiment. HFSS simulations were also verified using

equations given by T.H Lee in [17]. Thus, the experimental results, simulation results, and

results obtained from equations were in good agreement with each other. The simulation work

was thus used as a proof of concept for antenna design.

vi

TABLE OF CONTENTS

ACKNOWLEDGEMENT ...................................................................................................................iii ABSTRACT ..................................................................................................................................... iv LIST OF ILLUSTRATIONS............................................................................................................... x LIST OF TABLES ........................................................................................................................... xv Chapter Page 1. INTRODUCTION……………………………………..………..……………………… .................... 1

1.1 Importance of wireless Power…………………………………………………… . 3

1.2 Antenna characterization……………………………………………………… ..... 3

1.2.1 Antenna efficiency………………………………………… .................. 4

1.3 Contemporary antennas used for medical implants……..….......... ................. 5

1.4 Present work………………………………………………………………… .......... 6

1.5 Thesis indexing………………………………………… ...................................... 8

2. PLANAR SPIRAL ANTENNAS……………………………………….……………………. ......... 9

2.1 Introduction……………………………………………………… .......................... 9

2.2 Different types of spiral antennas………………………………… ................... 10

2.2.1 Equiangular spiral antenna……………………………… ............... 10

2.2.2 Archimedean spiral antenna………..................................... ........ 11

2.2.3 Cavity-backed slot antenna………………………… ..................... 12

2.3 Formulas used for planar spiral antennas…………………………… .............. 13

2.3.1 Modified wheeler formula……………………………… ................. 14

2.3.2 Expression based on current sheet approximation… .................. 14

vii

2.3.3 Data fitted monomial expression………………… ........................ 15

2.4 Planar spiral antenna fabrication techniques…………………………….. ....... 16

2.4.1 Photolithography process………………………… ........................ 16

2.4.1.1 Advantages of photolithography process ............... 18

2.4.1.2 Disadvantages of photolithography process ........... 18

2.4.2 Copper tape method…………………………… .......................... 19

3. ANSOFT HFSS SIMULATION……………………………...……… ............................... 21

3.1 Finite element analysis……………………………………………… .............. 21

3.2 Ansoft HFSS simulation………………………………………… .................... 22

3.3 Sample planar spiral antenna simulation setup………………… ................. 23

3.4 Matching simulations with equations …………………………….................. 24

3.5 Matching simulations with publications …………………… ......................... 27

3.6 Matching simulations with experiments…………………… ......................... 28

3.7 Matching simulations with bent antenna………………………… ................. 32

4. FLEXIBLE ANTENNA EXPERIMENTS……………………..……………… .................. 34

4.1 Class – E power amplifier circuit…………………………………… .............. 34

4.2 Introduction to experiments……………………………………… .................. 36

4.3 Antenna characterization setup requirements………………....................... 37

4.4 Experimental setup design…………………………………… ....................... 38

4.5 Experiment parameters……………………………………… ........................ 41

4.6 Antenna characterization results…………………………… ......................... 42

4.7 Explanation of the results……………………………………… ..................... 64

5. CONCLUSION………………………………………………………… ............................. 67

6. FUTURE WORK……………………………………………………………… ................... 69

6.1 Agilent module………………………………………… .................................. 69

6.2 Stacked spiral antennas………………… .................................................... 70

viii

APPENDIX

A. MATERIAL DATASHEETS AND SPECIFICATIONS……………… ............................. 71

A1. KAPTON DATASHEET………………………………… ............................... 72

A2. JVCC CFL-5CA COPPER FOIL TAPE SPECIFICATIONS… .................... 74

B. MATLAB PROGRAMS……………………………………………… ............................... 75

B1. CURRENTSHEET AND WHEELER FORMULAE ..................................... 76

B2. 3D PATTERN MATLAB PROGRAM .......................................................... 77

REFERENCES……………………………………………………………………… ........................ 78

BIOGRAPHICAL INFORMATION………………………………………………………………. ..... 81

ix

LIST OF ILLUSTRATIONS

Figure Page 1.1 mHealth platform……………………………………………………………………………… ..... 2 1.2 Fabricated planar spiral antenna………………………………………….………………… ..... 7 2.1 Equiangular spiral antenna……………………………………………….………………… ..... 10 2.2 Archimedean spiral antenna……………………………………………………………….. ..... 11 2.3 Cavity backed slot antenna………………………………………………………………… ..... 12 2.4 (a) Square, (b) Octagonal, (c) Hexagonal and (d) Circular planar spiral antennas [17]………………………………………... ............. 13 2.5 Mask for photolithography process………………………………………………………... ..... 17 2.6 Fabricated antenna using photolithography process………………………………………………………….……………………………. ..... 17 2.7 Comparison between photolithography fabricated and simulation model ………………………………………………………. ........... 18 2.8 Kapton taped on wooden platform …………………………...…………………………… ..... 19 2.9 Copper strips stuck on kapton substrate ………………………………………………… ..... 19 2.10 Mask taped on copper layer ……………………………………………………………… ..... 20 3.1 Simulation illustration …………………………………...………………………………….. ..... 22 3.2 (a) Top view and (b) Side view of simulation model.. ........................................................ 24 3.3 HFSS spiral antenna model……………………………………………………………...… ..... 25 3.4 Simulation result for 500 kHz frequency …………………………………………………. ..... 25

x

3.5 Simulation result for 5 MHz frequency ……………………………………………………. ..... 26

3.6 Simulation result for 500 kHz frequency………………………………………………….. ..... 26 3.7 Simulation result for 5 MHz frequency…………………………………………………… ...... 27 3.8 Publication [24] Vs. Simulation result …………………………………...………………... ..... 27 3.9 Publication [16] Vs. Simulation result …………………………………..………………... ...... 28 3.10 Spiral antenna fabricated on a wooden substrate …………………………..………… ...... 29 3.11 Comparison result between simulation and coil fabricated on wooden substrate …………………………………..…………………….. ...... 29 3.12 Spiral antenna fabricated on kapton substrate ………………………………..………. ...... 30 3.13 Comparison result between simulation and coil fabricated on kapton substrate ………………………………………………………..… ...... 30 3.14 Spiral antenna on wooden substrate with n=14, Lmw = 2.2388 µH, Lgmd = 2.5285 µH …………………………………….………. ...... 31

3.15 Comparison result between 14 turn coil and corresponding simulation model…………………………………..……………….. ...... 31 3.16 Comparison result for n=12, Lmw = 15.5567 µH,

Lgmd = 16.2004 µH …………………………………………………………………..…... ....... 31

3.17 Top view of bent antenna model………………………………………………….…….. ....... 32

3.18 Side view of bent antenna model………………………………….……………………. ....... 32

3.19 Comparison result at 1.3 MHz between simulation and experiment………………………………………………………………. ....... 33

4.1 Class – E amplifier circuit………………………………………………………...………. ....... 34

4.2 Top view of entire setup with a schematic of rotation angle of reader antenna ………………………………...……………………… ....... 38 4.3 Reconfigurable setup schematic ……………………………………………………...… ....... 39 4.4 Antenna misalignment, Antenna pattern, Cross section pattern using steps – 1, 2, 3…………………………………………….. ....... 39 4.5 3D pattern setup design……………………………………………………………..…… ....... 40 4.6 (a) Top view, (b) Side view of setup, (c) Back view of pvc where reader antenna is attached, (d) Top view of 3D setup, (e) Top view of 3D setup, (f) Front

xi

view of tag antenna plate…………………………………………………………………... ..... 40 4.7 Antenna pattern distance comparison for A1 with flat bend…………………………………………………….…………………. ....... 43 4.8 Cross-section pattern distance comparison for A1 with flat bend …………………………………………………………..…………………… ...... 43 4.9 Antenna misalignment distance comparison for A1 with C3 bend…………………………………………………………... ...... 44 4.10 Antenna pattern distance comparison for A1 with C3 bend……………………………………………………………………………..………. ...... 44 4.11 Cross-section pattern distance comparison for A1 with C3 bend……………………………………………………………………………. ...... 45 4.12 Antenna misalignment distance comparison for A1 with C2 bend………………………………………………………………………….… ...... 45 4.13 Antenna pattern distance comparison for A1 with C2 bend……………………………………………………………………..…. ...... 46 4.14 Cross-section pattern distance comparison for A1 with C2 bend………………………………………………………………………..………. ...... 46 4.15 Antenna misalignment distance comparison for A1 with C1 bend…………………………………………………….………………………….. ...... 47 4.16 Antenna pattern distance comparison for A1 with C1 bend……………………………………………………………………………….…….. ...... 47 4.17 Cross-section pattern distance comparison for A1 with C1 bend…………………………………………………………..…………………… ...... 48 4.18 Antenna misalignment curvature comparison for A1 at 4cm distance……………………………………………………………………………... ..... 48 4.19 Antenna misalignment curvature comparison for A1 at 6 cm distance…………………………………………………………………………….. ..... 49 4.20 Antenna misalignment curvature comparison for A1 at 8 cm distance…………………………………………………………………………….. ..... 49 4.21 Antenna pattern curvature comparison for A1 at 4 cm distance……………………………………………………………………….…….… ... 49 4.22 Antenna pattern curvature comparison for A1 at 6 cm distance………………………………………………………………………………....…… ... 50 4.23 Antenna pattern curvature comparison for A1 at 8 cm distance………………………………………………………………………………… ............ 50

xii

4.24 Cross-section pattern curvature comparison for A1 at 4 cm distance……………………………………………………………………………… . 50 4.25 Cross-section pattern curvature comparison for A1 at 6 cm distance…………………………………………………………………………… ..... 51 4.26 Cross-section pattern curvature comparison for A1 at 8 cm distance .................................................................................................................. 51 4.27 3D pattern for A1 at 4 cm distance.................................................................................. 52 4.28 3D pattern for A1 at 6 cm distance.................................................................................. 52

4.29 3D pattern for A1 at 8 cm distance.................................................................................. 53

4.30 Antenna pattern distance comparison for A2 with flat bend……………………………. ... 54

4.31 Cross-section pattern distance comparison for A2 with flat bend……………………………………………………………………………………..... ...... 54 4.32 Antenna misalignment distance comparison for A2 with C3 bend……………………………………………………….……………… ...... 55 4.33 Antenna pattern distance comparison for A2 with C3 bend……………………….… ...... 55 4.34 Cross-section pattern distance comparison for A2 with C3 Bend……………………………………………………………………………………….. ..... 56 4.35 Antenna misalignment distance comparison for A2 with C2 bend………………………………………………………………………………………... ...... 56 4.36 Antenna pattern distance comparison for A2 with C2 bend………………………………………………………………………………………... ...... 57 4.37 Cross-section pattern distance comparison for A2 with C2 bend………………………………………………………………………………………... ...... 57 4.38 Antenna misalignment distance comparison for A2 with C1 bend………………………………………………………………………………………. ........ 58 4.39 Antenna pattern distance comparison for A2 with C1 bend………………………………………………………………………………………... ...... 58 4.40 Cross-section pattern distance comparison for A2 with C1 Bend……………………………………………………………………………………….. ...... 58 4.41 Antenna misalignment curvature comparison for A2 at 4 cm distance…………………………………………………………………………….... .... 59 4.42 Antenna misalignment curvature comparison for A2 at 6 cm distance…………………………………………………………………………….... .... 59

xiii

4.43 Antenna misalignment curvature comparison for A2 at 8 cm distance…………………………………………………………………………….... .... 60 4.44 Antenna pattern curvature comparison for A2 at 4 cm distance…………………………………………………………………………….... .... 60 4.45 Antenna pattern curvature comparison for A2 at 6 cm distance…………………………………………................................................ ....... 60 4.46 Antenna pattern curvature comparison for A2 at 8 cm distance………………………………………………… ............................................ 61 4.47 Cross-section pattern curvature comparison for A2 at 4 cm distance…………………………………………………………………………….... .... 61 4.48 Cross-section pattern curvature comparison for A2 at 6 cm distance…………………………………………………………………………….... .... 61 4.49 Cross-section pattern curvature comparison for A2 at 8 cm distance…………………………………………………………………….............. ..... 62 4.50 3D pattern for A2 at 4 cm distance……………………………………………………… .... 62 4.51 3D pattern for A2 at 6 cm distance……………………………………………………… .... 63 4.52 3D pattern for A2 at 8 cm distance…………………………. ........................................... 63 6.1 Agilent module[12] ………………………………………................................................... . 69 6.2 Stacked spiral antenna [10]…………………………………………................................. .. 70

xiv

LIST OF TABLES Table Page 2.1 Modified wheeler formula constants…………………………………………………………. ..... 14

2.2 Current sheet approximation formula constants…………………...………………………. ..... 15

2.3 Data fitted monomial expression constants …………………………….………………….. ..... 16

4.1 Experiment parameters………………………………..……………………………………….. ... 42

1

CHAPTER 1

INTRODUCTION

Healthcare to anyone, anytime and anywhere by removing locational, time and

other restraints is the vision of “Pervasive Healthcare” [1]. The introduction of wireless

technologies in healthcare environment has led to an increased accessibility to healthcare

providers, more efficient tasks and processes, and a higher overall quality of healthcare

services. Health monitoring can be used to monitor physical health, emotional health, and

behavioral health. Health monitoring involves measuring several medical parameters

simultaneously over a long term without interrupting the patient in performing his daily chores.

Measurements can include recording and digitization of vital signs like acid, non – acid reflux in

the esophagus, blood pressure, electrocardiogram (ECG), respiration rate, pulse, oxygen

saturation, gastro paresis etc. The monitoring system should transmit both routine vital signs

and alerting signals, when the obtained vital signs cross a particular threshold value.

Comprehensive health monitoring systems include monitoring using network systems such as

those networks which are implemented by using Adhoc wireless technology, personal area

networks implemented by Zigbee Technology, RFID technology, large area networks

implemented by Wi-Fi Technology, cellular networks etc. With the coverage and scalability

provided by wireless networks used in wireless health monitoring systems, a large coverage

which includes coverage through rural and urban areas covering both indoor and outdoor

environments can be provided. A new feature has been added to the ongoing revolution in

wireless health monitoring technologies and the term coined for this new feature is – mHealth.

2

mHealth or mobile health is a term used for usage of mobile communication devices such as

mobile phones, tablet computers, PDA for providing health services and information.

Figure 1.1 mHealth platform [31]

On the contrary, many challenges are being faced in providing better health services, such as a

significant number of medical errors, considerable stress on health care providers, an increased

cost of healthcare services and an exponential increase in the number of seniors and retirees in

developing countries. Worldwide, there has been an increase in the number of people suffering

from physical and cognitive diseases, and with the aging of world population, this number will

grow even more significantly in the future. About 37 million people (which is about 14% of the

world population) in the United States alone are suffering from some or the other form of

disability. About 40% of seniors in the United States are suffering from disabilities. It has been

shown that health monitoring can reduce the number of readmissions for patients suffering from

chronic health problems. To support the long – term health care needs of the patients,

comprehensive wireless health care monitoring solutions must be developed for homes, nursing

homes and hospitals. As a result of the comprehensiveness of such systems, many challenges

need to be taken care of in designing and developing such systems, including diverse

3

requirements of health monitoring, wireless and efficient powering of the body implants that are

used to monitor the patients, medical decision making on the health care needs of the patients,

patient comfort in wearing the powering device to power the body implant. In this thesis work

problems related to wireless efficient powering of body implant and patient comfort in wearing

the powering device – planar spiral antenna along with transmitter circuit are addressed by

characterizing a flexible antenna.

1.1 Importance of wireless power

Electronic devices can be worn around the body or implanted inside the body and

they can be used to perform a variety of prosthetic, therapeutic and diagnostic functions [2].

However efficiently powering the body implants is a problem, and researchers are laying a lot of

emphasis on studying and solving these problems. For a long time medical practitioners have

implanted body electronics inside the body which were powered by non – rechargeable

batteries. This involved repetitive surgeries to replace the batteries. In order to solve this

problem, wireless power transfer technology was used. There are several ways to transfer

wireless power to power the body implants - magnetic resonance coupling and capacitive

coupling.

In this thesis work, the body implant is powered using resonant magnetic coupling

phenomenon. An advantage of making use of the coupling phenomenon is that more than one

body implant can be powered using a single source.

1.2 Antenna characterization

The effectiveness with which body implants can be powered can be learned by

studying the radiation pattern of the device that is used to power the implant. In order to study

the radiation pattern of a device, one of the methods adopted is studying the efficiency pattern.

The efficiency pattern of the radiating device is studied by making use of a stable setup, so that

experiments performed are repeatable. Also the stability of the setup matters when the power

transferred is low in order to minimize voltage fluctuations at the receiver end. In this thesis

4

work, the setup was made by using thick acrylic boards and pvc pipes to ensure that antenna

bend is perfect, and results obtained are accurate.

1.2.1 Antenna efficiency

The definition of antenna efficiency inside a lossy media is not obvious, since the far – field

is attenuated [6]. The standard definition of antenna gain is G(θ,φ) = ῃD(θ,φ) where ῃ is the

efficiency factor, D(θ,φ) is the directivity of the antenna and is defined from the normalized

power pattern Pn as

2

2

( , )( , )( , )

( , ) ( , )

n

n average

average

FPD

P F

(1.1)

where ( , )F is the far-field amplitude. The normalized power pattern Pn as defined from the

Poynting vector, S = .S r , is

2

2max

max

( , )( , )( , )

( , ) ( , )n

FSP

S F

(1.2)

where the far-field amplitude is defined as

( , ) lim ( , , ) jkr

k rF E r kre

(1.3)

The far-field amplitude can also be expressed as

'( , ) ( ). ( )

s

jk r

sV

F j k I rr e J r dv (1.4)

where ( ')sJ r is the source current density in a volume Vs.

This definition also applies to antennas inside lossy media. The radiating power of an

antenna gets attenuated by matter as it travels inside a lossy medium. This has the

consequence that the position of the origin is important as it influences the shape of the pattern.

5

This is the reason why Antenna characterization experiments in this thesis have been

performed by using different orientations of the origins of both the antennas.

The definition of antenna efficiency inside a lossy medium has to be adopted rather than the

one used in air. The usual way of defining antenna efficiency is

radiatedlossless

accepted

P

P (1.5)

HereacceptedP is the power that is accepted by the transmitting antenna, i.e., the input power to

the antenna subtracted from the reflected power from the transmitting antenna. In the case of

an antenna in matter this definition has to be modified, as the quantity radiatedP will vary with the

radius r. The radiated power is dependent on distance between the transmitting and receiving

antenna.

P(r) = Poe-2Im[k]r

where r is the radius at which the power is calculated and k is the dispersion

constant. Antenna efficiency in a lossy matter is defined as

olossy

accepted

P

P (1.6)

where PO is the radiated power. The above definition is also valid for a lossless medium.

1.3 Contemporary antennas used for medical implants

The antennas used for powering medical implants are – dipole antenna, wire Antenna,

circumferential quarter wave, circumferential PIFA (planar inverted F antennas), patch and

magnetic coil antennas.

Wire antennas such as dipole antennas [4] cannot be used at operating frequencies of

1.3 MHz, because for them to operate efficiently and resonate at the frequency of interest, they

have to be half wavelength long. This is around 233 meters. Antennas of such size cannot be

used to power medical implants and hence are not feasible.

Circumferential quarter wave antennas are compact plates of circular shape. They have

high bandwidths, but their physical size is frequency dependent.

6

Circumferential PIFA and patch antennas are frequency dependent antennas. Their physical

size will change with change in frequency of interest. Moreover more time and effort is needed

to fabricate these antennas, and it is difficult to make them work properly at wide bandwidths.

Also the physical size of these antennas will be very large if they are designed for 1.3 MHz,

which is the frequency of interest in this thesis.

Magnetic coil antennas have the advantage that the resonant radiation frequency can

be controlled by using a resonant circuit instead of the physical size of the structure. One such

example of magnetic coil antennas is a spiral antenna.

1.4 Present work

In the present work, a special type of spiral antenna called flexible spiral antenna has

been used. There are many advantages of making use of flexible spiral antennas –

They are flexible; so they can bend to the shape of an object on which they are placed.

In the present work they are placed on the stomach, where they take the shape of the

stomach.

Spiral antenna coil turns are distributed across the radii [3], rather than concentrated on

the outer radius of the antenna. Zeirhofer and Hochmair (1996) studied the

enhancement of magnetic coupling between coils and found that the magnetic coupling

increases when the turns are distributes across the radii.

Spiral antennas are frequency independent. The physical antenna size need not have

to be altered if the frequency of interest changes. Changes have to only be made on the

resonating circuit to achieve high efficiency at the frequency of interest.

Spiral antennas used are planar and compact, and they can be mounted easily on the

surface of interest. They do not obstruct movement of the user and are called low

profile antennas.

They are economically cheap to fabricate and the failure rate of the fabrication process

is less than that of patch antennas. Also fabrication time is short.

7

The polarization, radiation pattern, and impedance of such antennas remain unchanged

for large bandwidths.

When the antenna shape is made to conform to the shape of the human stomach, the

antenna bends in a concave manner, this results in increase in efficiency, since all the

magnetic field is focused on the medical implant.

The bandwidth of operation of such antennas can be as high as 30:1; hence they can

be used for wideband applications.

Although the inductors used in the design have small Q values, their inductances are

defined over a broad range of process variations.

The spiral antennas that were fabricated in this work are square in shape, because it is easier to

fabricate antennas of this shape using the copper tape method instead of circular or trapezoidal

shaped antennas. Also these antennas can be easily modeled using the three dimensional

electromagnetic finite element based software called Ansoft HFSS. Using this software the

antenna inductance can be verified with experiments and equations written by T.H Lee in [17].

Figure 1.2 Fabricated planar spiral antenna

8

1.5 Thesis indexing

Chapter 1 deals with the issues faced in health monitoring, importance of wirelessly powering

medical implants, the different types of antennas used in wireless powering of implants and the

advantages of making use of flexible planar spiral antennas.

Chapter 2 deals with an introduction to spiral antennas, the different types of spiral antennas

used and the design formulas associated with planar spiral antennas. It describes the different

fabrication techniques used and the steps involved in fabrication, advantages and

disadvantages of techniques used.

Chapter 3 gives an introduction to finite element method (FEM) simulations, used by the 3D

electromagnetic simulation software – Ansoft HFSS. It describes the method involved in setting

up the simulation parameters and simulation model. It describes the different steps used to

verify the simulations i.e. using theoretical equations, publication results, flat antenna and bend

antenna experimental results.

Chapter 4 gives an introduction to the Class – E power amplifier circuit. It describes the typical

parameters that are to be considered for carrying out antenna characterization experiments, an

introduction to the design and fabrication of the stable setup, an introduction to the experimental

parameters, graphical results obtained by carrying out different types of experiments for

antenna characterization, and the explanation of the results obtained.

Chapter 5 gives the conclusion of the Antenna characterization experiments.

Chapter 6 describes the method used to carry out future antenna characterization experiments.

It also describes the method used to increase the range of power transfer.

9

CHAPTER 2

PLANAR SPIRAL ANTENNAS

2.1 Introduction

The numerous applications of electromagnetics to the advances of technology

have necessitated the exploration and utilization of most of the electromagnetic spectrum [22].

In addition, the advent of broadband systems, have demanded the design of broadband

radiators. The use of simple, small, lightweight, and economical antennas designed to operate

over the entire frequency band of a system, would be most desirable. Although in practice all

the desired features and benefits cannot easily be derived from a single radiator, most can

effectively be accommodated. Previous to the 1950’s, antennas with broadband pattern and

impedance characteristics had bandwidths not greater than 2:1. In the 1950’s, a breakthrough in

the antenna revolution was made which extended the bandwidth to as great as 40:1 or more.

The antennas introduced by the breakthrough were referred to as frequency independent, and

they had geometries that were specified by angles between spiral arms. One such example of

frequency independent antennas is planar spiral antennas.

Planar spiral antennas are viable candidates for near field wireless power transmission

to the next generation of high performance medical implants with extreme size constraints. So

far, antenna coils for powering medical implants have been fabricated with thin wires using

sophisticated winding machinery [16]. However, the need for much smaller footprints in the next

generation of high performance implantable devices calls for higher geometrical precision and

potential for integration on chip or on package. This would require different fabrication

techniques that would result in well-defined planar structures. Also these antennas offer more

flexibility in defining their characteristics. A simple change in their shape could change the

10

polarization pattern. Planar antennas of different shapes are being studied to investigate which

shapes are best for powering medical implants.

2.2 Different types of spiral antennas

There are different types of spiral antennas such as equiangular spiral antennas or log periodic

spiral antennas, archimedean spiral antennas and cavity backed slot antennas [21]. These will

be discussed in the following sections.

2.2.1 Equiangular spiral antenna

The log - periodic spiral antenna or the equiangular spiral antenna has each arm defined

by the polar function a

or R e where oR is a constant that controls the initial radius r of the

spiral antenna. The parameter a controls the rate at which the spiral antenna flares or grows as

it turns. The above equation states that the spiral antenna radius grows exponentially as it turns.

A plot of log – periodic spiral antenna is shown below

Figure 2.1 Equiangular spiral antenna [31]

The planar spiral antenna of Figure 2.1 will have peak radiation directions into and out of the

screen (broadside to the plane of the spiral, in both the front and the back). The antenna will

radiate right hand circularly polarized fields out of the screen, and left hand circularly polarized

fields into the screen. The parameters that affect the radiation of the spiral antenna include:

1. The outer radius (Rspiral) – This parameter determines the lowest frequency of operation

of the spiral antenna.

11

2

low

low spiral

c cf

R (2.1)

2. The flare rate (a) – The rate at which the spiral grows with angle is the flare rate. If it is

too small, the spiral is tightly wrapped around itself.

3. Feed structure – The feed must be controlled with a balun so that the spiral has

balanced currents on either arm. Also, the highest frequency in the spiral antenna’s

operating band depends on the innermost radius of the spiral Ro.

4

Upper

Upper o

c cf

R (2.2)

4. Number of turns (N) – This is also a design parameter for the spiral antenna.

Radiation occurs from the spiral antenna when the current in the spiral arms are in phase.

2.2.2 Archimedean spiral antenna

Each arm of the Archimedean spiral is defined by the equation r a . This equation

states that the radius r of the antenna increases linearly with the angle . The parameter a, is

simply a constant that controls the rate at which the spiral flares out. The second arm of the

archimedean spiral is the same as the first, but rotated 180 degrees. A plot of the Archimedean

spiral is shown below

Figure 2.2 Archimedean spiral antenna [31]

12

2.2.3 Cavity-backed slot antenna

The antennas discussed till now are those that are planar slot antennas. They can be

used in metallic backed objects such as aircraft body etc. It is desirable for the slot antenna to

be cavity backed, because this isolates the antenna from what is behind it, so that it can be

mounted on objects without worrying about retuning the antenna. These antennas come in two

types.

The first is a simple metallic backing separated from the antenna by some distance d.

This metallic backing will cause reflections of the radiated fields that enter the cavity, and these

often cancel the radiated fields that travel broadside to the plane of the antenna. If the depth d is

an integer multiple of half wavelength, the reflected field will tend to cancel the forward travelling

field of the spiral antenna, thus leading to poor radiation. Hence the wideband characteristics of

a metallic cavity backed slot antenna are less. An alternate method is to make use of absorbing

material to cavity back the spiral antenna. In this case, the reflected fields from the cavity will be

attenuated, so that there is no destructive interference, so the antenna will maintain its

wideband characteristics. This however tends to decrease the efficiency of the spiral antenna,

since roughly half the radiated power is radiated. However, this loss of efficiency is 3 dB, which

is often tolerable. An example of the cavity backed slot antenna is the cavity backed

Archimedean spiral antenna is shown below.

Figure 2.3 Cavity backed slot antenna [31]

13

2.3 Formulas used for planar spiral antennas

There are some formulas that accurately express the dc inductance of square, hexagonal,

octagonal, and circular spiral inductors [17]. The accuracy of these expressions was modeled

using field solver simulations, ASITIC, and using measurement data. For a given shape, an

inductor is completely specified by the number of turns, n, the turn width w, the turn spacing s,

the outer diameter dout, the inner diameter din, the average diameter 0.5( )avg out ind d d and

the fill ratio defined as( )

( )

out in

out in

d d

d d

. The thickness of the inductor has a meager effect on

the inductance and hence is ignored. The expressions for inductance are defined in [17].

(a) (b)

(c) (d) Figure 2.4 (a) Square, (b) Octagonal, (c) Hexagonal and (d) Circular planar spiral antennas [17]

14

2.3.1 Modified wheeler formula

Wheeler presented several formulas for planar spiral inductors, which were intended for

discrete inductors. A simple modification of the wheeler formula can be used to obtain an

expression that is valid for planar spiral integrated inductors.

2

1

21

avg

mw o

n dL K

k

(2.3)

where is the fill ratio defined, previously. The coefficients 1K and 2K are layout dependent

and are shown in Table 2.1. The ratio, ,

is the fill ratio and represents how hollow the inductor

is. For a smaller ratio the inductor is hollow, and for a larger ratio the inductor is full. Two

inductors with the same average diameter but with different fill ratios will have different

inductances. The full inductor has a smaller inductance because its inner turns are closer to the

center of the spiral and hence would contribute less positive mutual inductance and more

negative mutual inductance

Table 2.1 Modified wheeler formula constants

2.3.2 Expression based on current sheet approximation

Another simple and accurate expression for planar spiral inductances can be obtained by

approximating the sides of the spirals by symmetrical current sheets of equivalent current

densities. For example, in case of a square, 4 identical current sheets are obtained. The current

sheets on opposite sides are parallel to each other, whereas the adjacent ones are orthogonal

to each other. Using symmetry and the fact that orthogonal current sheets have zero mutual

inductance, the computation of the inductance is now reduced to evaluating the self-inductance

of one sheet and the mutual inductance between the opposite current sheets. The self and

15

mutual inductances are evaluated using the concepts of geometric mean distance, arithmetic

mean distance, and the arithmetic mean square distance. The resulting expression [17] is

2

1 223 4(ln( ) )

2

avg

gmd

n d c cL c c

(2.4)

where ci are layout dependent given by the values in Table 2.2, s and w are the spacing

between the inductor turns and width of the inductor turns. Although the accuracy of the

expression reduces as the ratio, s/w, becomes large, it exhibits a maximum error of 8 % for s ≤

3w.

Table 2.2 Current sheet approximation formula constants

2.3.3 Data fitted monomial expression

This expression for inductance is obtained using a data fitting technique. The expression

is given by (2.5).

3 51 2 4

outmon avgL d d n s

(2.5)

where the coefficients β and αi are layout dependent and given in Table 2.3. The above

expression is monomial in the variables , , ,out avgd d n and s. The coefficients in (2.5) were

obtained using the following expressions 1x = log outd , 2 logx ,3 log avgx d , 4 logx n ,

5 logx s . Using logarithms, the expression (2.5) can be expressed as (2.6).

1 1 2 2 3 3 4 4 5 5log oy L x x x x x (2.6)

16

where logo . This is a linear plus constant model of y as a function of x, and is easily fit

by various data fitting and regression techniques. To develop the models, a simple least

squares fit is used. Also i is chosen to minimize (2.7).

( ) ( ) ( ) ( ) ( ) ( ) 2

1 1 2 2 3 3 4 4 5 5

1

( )N

k k k k k k

o

k

y x x x x x

(2.7)

where the sum is over a family of 19000 inductors.

Table 2.3 Data fitted monomial expression constants

The monomial expression (2.5) is useful since, like the other expressions (2.3), (2.4), it is

accurate and simple. It is used in optimal design of inductors

2.4 Planar spiral antenna fabrication techniques

There are different methods adopted in this research to fabricate the planar spiral

antennas. However, after making use of many methods, copper tape method was found to be

the easiest. The methods used are listed below.

2.4.1 Photolithography process

The steps used in fabricating the planar antenna using the photolithography process are

as follows

a. The edges of a desensitized double sided copper clad board are smoothened

using an electric razor blade.

b. The mask is prepared using the AutoCAD program and printed on the rough

side of a plastic transparency. This is placed on the photoresist coated copper

clad board.

17

c. A sheet of Plexiglas is placed on the mask and the board, ultra violet light is

used to expose the sensitized board with the mask for 10-15 minutes.

d. The exposed part of the photoresist is then washed away using potassium

hydroxide solution in a beaker.

e. A 1:1 mixture of 30% hydrogen peroxide solution and hydrochloric acid is used

to etch the copper from the board.

f. Acetone is then used to remove the unexposed photoresist. The planar spiral

antenna is then ready to be used.

Figure 2.5 Mask for photolithography process

Figure 2.6 Fabricated antenna using photolithography process

18

2.4.1.1 Advantages of photolithography process

The process is less time consuming. It takes about half an hour to complete the antenna

fabrication process. The precision achieved is high as compared to other fabrication techniques

2.4.1.2 Disadvantages of photolithography process

The process is relatively expensive, because we are making use of different chemicals,

FR4 substrate. The process is very smelly, because of the release of hydrochloric acid vapors

during the etching process, which are corrosive in nature.

Since FR4 substrate can be used only once for antenna fabrication, different antenna

designs can only be executed using different substrates. This technique was discarded for cost

reasons.

Figure 2.7 Comparison between photolithography fabricated and simulation model

Figure 2.7 shows a plot of inductance vs. frequency for a 7 turn coil with a width (w) = 5 mm and

spacing between turns (s) = 5 mm fabricated using photolithography process. The external

diameter of the coil (dout) = 155 mm and internal diameter (din) = 10 mm. The coil inductance is

compared to the inductance of a coil with same parameters modeled using Ansoft HFSS

electromagnetic simulation software that makes use of the finite element method to perform

simulations.

19

2.4.2 Copper tape method

Using this method, we can make use of a single substrate and easily change the design

of the antenna by peeling the copper tape and using another tape with a different design. The

steps followed to fabricate planar spiral antennas using this method are:

The substrate is first made stable by fixing it in a table using tape. The substrate used in

this research is kapton. Before making use of kapton substrate, wood was used for

preliminary experimental work.

Figure 2.8 Kapton taped on wooden platform

Copper tape is then taped on top of the substrate. Since the tape is of a particular

width, many strips of it are taped to cover the entire substrate. The tapes are attached

such that they partially overlap. This overlapping can cause discontinuities which can

later be prevented by using solder.

Figure 2.9 Copper strips stuck on kapton substrate

20

The mask of the planar spiral antenna made using AutoCAD software [18, 19] is then stuck on the copper tape.

Figure 2.10 Mask taped on copper layer

An impression is made on the mask along the spiral boundary using a pen knife and

then the mask is removed.

The copper tape is peeled from the substrate to obtain the planar spiral antenna.

The discontinuities in the coil are then soldered using tin-lead wire.

The antenna is then tested using an impedance analyzer. If the analyzer shows a

negative inductance, the antenna is checked for discontinuities and appropriate

connections are made until the analyzer reads positive inductance.

21

CHAPTER 3

ANSOFT HFSS SIMULATION

3.1 Finite element analysis

The finite element method (FEM) is a powerful numerical analysis technique which is

well-suited to and appropriate for solving a wide variety of electromagnetic applications

problems computationally [29]. Amongst the many methods used in computational

electromagnetics, its ability to manage problems with complex geometries, as well as its broad

applicability to static, quasi-static, wave and transient systems, and to problems containing

material regions that are nonlinear, inhomogeneous and anisotropic, all make the FEM one of

the most versatile and powerful computational analysis and simulation schemes available today.

Moreover, the solid theoretical foundations on which the FEM is based, as well as the rigorous

mathematical analysis concerning the existence, convergence, and the uniqueness of finite

element solutions that have been established, further justify its use in electromagnetics

research and design.

While FEM methods are presently used extensively for electromagnetics analysis and

design, the use of adaptive FEM methods has gained considerable attention in recent years

from numerical analysts for solving problems more efficiently than standard FEM methods

permit. The accuracy of a finite element solution is directly dependent on the number of free

parameters used to mathematically represent the problem, and how effectively those

parameters, or mathematical DOF(degrees of freedom), are distributed over the problem space.

Consequently, the most efficient distribution of DOF for a problem is that which provides a

sufficiently accurate solution for the lowest number of free parameters.

Currently, the only practical way to achieve this objective is by using adaptive solution

strategies which are capable of intelligently evolving and improving an efficient distribution of

22

DOF over the problem domain by establishing solution error distributions, and then adjusting or

adding DOF to the discretization to correct them. By increasing the number of DOF in the

vicinities of higher solution error only, it is possible to make the most significant improvement in

the global accuracy of the finite element solution, for the minimum additional computational

cost.

3.2 Ansoft HFSS simulation

High frequency Structural Simulator (HFSS) is a high performance full wave

electromagnetic simulator [13] that employs the finite element method (FEM), adaptive

meshing, and brilliant graphics to give insight to 3D electromagnetic problems. Design creation

using Ansoft HFSS involves the following -

Parametric model generation – The geometry of the model is created, the boundaries

and excitations are then defined.

Analysis setup – The solution setup and the frequency sweeps are then defined.

Results – 2D reports and field plots are then obtained.

Solve Loop – The solution process is fully automated.

Figure 3.1 is an illustration showing the simulation procedure in Ansoft HFSS.

Figure 3.1 Simulation illustration

23

3.3 Sample planar spiral antenna simulation setup

The following steps are involved in creating and simulating a planar spiral antenna -

To create a new project, we go to File and click on new project. From the project menu,

insert HFSS design option is selected.

Then solution type option from the HFSS menu item is selected and driven terminal

solution type is clicked.

From the 3D modeler menu item, unit’s option is selected and the appropriate units are

selected.

The 3D model of planar spiral antenna is then created using the HFSS graphical user

interface. Boundary conditions are applied and internal ports (lumped ports) are

assigned to the model. Care is taken that appropriate material properties are assigned

to different parts of the model.

From the HFSS menu option, analysis setup is selected and Add solution setup sub

option is then selected. The solution frequency, the maximum number of passes, and

the maximum error between each pass is then defined.

From the HFSS menu option, Analysis Setup is selected and Add sweep sub option is

then selected. The start frequency, stop frequency and the sweep type which is usually

interpolating is then selected.

The file is then saved and the validation check option from the HFSS menu option is

then selected to ensure whether there are any errors or not in the simulation setup.

The Analyze All option from the HFSS menu option is then selected. The simulation

then begins and ends when the solution converges.

From the HFSS menu option, the results option is selected and the create report sub

option is then selected.

24

The formulas for inductance and quality factor for the antenna are then inputted.

11

1Im( )

(2* * )

YL

pi freq & 11

11

1Im( )

1Re( )

YQ

Y

. These formulas [15, 28] are in the y axis of the

curve that is plotted with the x axis as frequency. The results are then exported in an

excel format and the values are obtained for frequency of interest.

A sample planar spiral inductor modeled using Ansoft HFSS is shown below –

(a) (b)

Figure 3.2 (a) Top view and (b) Side view of simulation model

3.4 Matching simulations with equations

Different designs of planar antennas were implemented in order to confirm the validity of

the equations in [17] with the simulation results. Since, it is clearly mentioned in [17] that the

25

inductance equations are only valid for dc frequencies; antennas with different designs were

simulated for different number of turns and for different dc (less than 800 kHz) frequencies.

Shown in Figure 3.5 is an antenna with a turn spacing and width of 8 mm, thickness of 0.1 mm

and an external diameter of 15 mm. An FR4 substrate of size 30 mm x 30 mm was used in

simulation.

Figure 3.3 HFSS spiral antenna model

Simulations were carried out for 500 KHz, 1 MHz, 1.3 MHz, 5 MHz and 10 MHz frequencies.

The values Lmw and Lgmd are inductance values obtained from formulas in [17]. The value Lhfss is

obtained from simulations performed in this research. As the frequency increases, it is observed

that the deviation of the three inductances increases from each other. This validates the fact,

that the inductance values Lmw and Lgmd are only valid for dc frequencies.

Figure 3.4 Simulation result for 500 kHz frequency

26

Figure 3.5 Simulation result for 5 MHz frequency

Shown in Figure 3.6 is another design of an antenna where the turn spacing and width are

equal to 0.2 mm and the thickness is equal to 0.1 mm. The external diameter of the antenna is

50 mm. The substrate size is 100 mm x 100 mm and thickness is 1.5 mm. The graphs shown in

Figure 3.8 and Figure 3.9 also show the deviation of the three inductances with the increase in

frequency.

Figure 3.6 Simulation result for 500 kHz frequency

27

Figure 3.7 Simulation result for 5 MHz frequency

3.5 Matching simulations with publications

In order to ensure that the simulations performed using Ansoft HFSS are correct,

Several IEEE papers [25, 26, 27] were implemented and the HFSS results obtained from

simulations carried out in research work were matched with the HFSS results obtained from the

IEEE papers. Figure 3.10 shows the results obtained from [24].

Figure 3.8 Publication [24] Vs. Simulation result

Here the values Lmy Simulation and Lpaper are the inductances of the model implemented in this

research and the author. There is a small variation in the results between the simulation by

28

these authors and this research simulation, this can be explained by the different software

versions used and the size of the radiation boundaries.

HFSS simulations were also implemented to match results in another paper [16]. The

graphs showing research results and those in [16] are shown in Figure 3.11. This time the

results being matched are the quality factor of the coils.

Figure 3.9 Publication [16] Vs. Simulation result

Here the values Qmy simulation and Qpaper are the quality factors obtained from HFSS

implementation and the implementation by [16]..

3.6 Matching simulations with experiments

The HFSS simulations were later matched with experiments. Different experiments were

run using antennas of different design and on different substrates. The results of which are

shown below. A 7 turn antenna of thickness 0.035 mm, turn and spacing width 5 mm is

fabricated on a 13.5 mm thick wooden substrate using the copper tape method. The Lmw and

Lgmd values of this antenna are 3.384 and 3.4975 micro-henry respectively. A picture of the

antenna is shown below. The value Lexp of the antenna is obtained after subtracting the probe

inductance from the inductance obtained using the LCR meter.

29

Figure 3.10 Spiral antenna fabricated on a wooden substrate

Figure 3.11 Comparison result between simulation and coil fabricated on wooden substrate

Antenna with the Figure 3.12 design is then fabricated on kapton substrate using Copper tape

method.

30

Figure 3.12 Spiral antenna fabricated on kapton substrate

Figure 3.13 Comparison result between simulation and coil fabricated on kapton substrate

A 14 turn planar spiral antenna is then fabricated on a wooden substrate of thickness

13.5 mm. The turn and spacing width of the antenna is 5 mm and the thickness of the copper

coil is 0.035 mm. The fabrication is performed using copper tape method. HFSS simulations

and measurements are then performed for different turns and the results are compared.

31

Figure 3.14 Spiral antenna on wooden substrate with n=14, Lmw = 2.2388 µH, Lgmd = 2.5285 µH

Figure 3.15 Comparison result between 14 turn coil and corresponding simulation model

Figure 3.16 Comparison result for n=9, Lmw = 6.8542 µH, Lgmd = 7.1112 µH

32

3.7 Matching simulations with bent antenna

The coil antenna is fabricated on a flexible substrate in order to provide comfort to the patient

who is wearing it. Since a bent coil is used, simulation has to be also performed in HFSS for

bent coils. Since HFSS does not provide a method to bend the coil, the coil model has to be

exported to a format that can be used by a different program that allows for a bent substrate to

bend it. AutoDesSys Inc. provides software called Bonzai 3d that can be used to bend the coil

along any plane and to any degree. After bending the coil, the bent coil model can be exported

to a format that can be imported into HFSS for performing simulations.

After modeling the coil in HFSS, the model can be exported in .sat format. The (.sat)

file can later be imported into Bonzai 3d software [20].The software has a bend function that can

be used to bend the coil antenna to any degree and along XY, YZ or XZ axis. In this case the

model was bent along XZ axis. The bent coil model is then exported in .sat (standard ACIS text)

format. The file is then imported into HFSS and after applying boundary conditions, excitations,

the simulation is then started. A sample 7 turn coil model is simulated after being bent and the

results are compared to experimental results.

Figure 3.17 Top view of bent antenna model Figure 3.18 Side view of bent antenna model

33

Figure 3.19 Comparison result at 1.3 MHz between simulation and experiment

34

CHAPTER 4

FLEXIBLE ANTENNA EXPERIMENTS

The efficiency pattern of the flexible antenna is obtained using a Class – E power amplifier

circuit [5, 9, 11], a receiver circuit, and a well, designed setup which ensures the stability and

repeatability of all experiments conducted using the setup.

4.1 Class-E power amplifier circuit

Figure 4.1 Class – E amplifier circuit

35

Class – E amplifiers can operate with power losses smaller by a factor of 2.3 [11] compared

with conventional Class – B or Class - C amplifiers making use of the same transistor at the

same frequency and output power. Class – E power amplifiers also have a very high efficiency

compared to other conventional amplifiers. Hence, it is used in this experimental work described

in this thesis. Also the MOSFET used is IRF 530 (enhancement type N channel MOSFET),

because it has a low threshold voltage and has a higher efficiency as compared to IRF 510

transistor.

The amplifier operates as follows. The 100 µH inductor connected at the drain acts as a

constant current source. When the switch closes, LReader supplies current through the switch and

no current flows through CDS. When the switch opens, the inductor supplies current through CDS

and no current flows through the switch thereby building a voltage across the switch. The

amplifier operation is such that the voltage and current at the switch have a phase difference of

90 degrees at all times. This results in very low power losses and high efficiency of the

amplifier. There is a particular value of CReader and CTag at which the amplifier resonates at 1.3

MHz frequency and the efficiency of power transfer between the LReader and LTag through the

resonant magnetic coupling phenomenon, is maximum. Care is always taken that resonance is

achieved before obtaining the efficiency patterns for any of the antenna characterization

configurations so that stability and hence repeatability of the experiments is ensured.

The efficiency of power transfer between the two coils is calculated based on equations (4.1) –

(4.3).

i DD DDP V I (4.1)

where iP is the input power and the dc voltage DDV is fixed at 4 V.

2 2

500o

Load

V VP

R (4.2)

36

where V is the voltage across the load resistor and oP is the output power delivered to the

load.

The efficiency of power transfer is then calculated using (4.3).

In % = 100%O

i

P

P . (4.3)

4.2 Introduction to experiments

Four different experiments are performed to characterize the flexible antenna.

Antenna misalignment – The tag antenna is fixed and the reader antenna is rotated

along a particular curvature. The angle at which both the reader antenna and tag

antenna are center aligned is considered 0 degrees. The clockwise rotation of the

reader antenna is considered positive rotation and the anticlockwise rotation of the

reader antenna is considered negative rotation. A graph is obtained with the reader

antenna angular rotation plotted along the x axis and the efficiency values at the

corresponding reader antenna positions plotted along the y axis. The experiment is

repeated for all 3 different curvatures of the reader antenna and at tag antenna

distances of 4 cm, 6 cm and 8 cm from the reader antenna.

Antenna pattern - The reader antenna is fixed and the tag antenna is rotated along its

axis. The angle at which both the reader antenna and tag antenna are center aligned is

considered 0 degrees. The clockwise rotation of the tag antenna is considered positive

rotation and the anticlockwise rotation of the tag antenna is considered negative

rotation. A graph is obtained with the tag antenna angular rotation plotted along the x

axis and the efficiency values at the corresponding tag antenna positions plotted along

the y axis. The experiment is repeated for all 3 different curvatures of the reader

antenna and at tag antenna distances of 4 cm, 6 cm and 8 cm from the reader antenna.

Cross section pattern – The reader antenna is fixed and the tag antenna is moved

upwards and downwards with the distance between the two antennas fixed. The

37

position at which both the reader antenna and tag antenna are center aligned is

considered the origin. The upward movement of the tag antenna is considered negative

movement from the origin and the downward movement of it is considered positive

movement from the origin. A graph is plotted with tag antenna movement from the

origin, along the x axis and the efficiency values at the corresponding tag antenna

positions along the y axis. The experiment is repeated for all 3 different curvatures of

the reader antenna and at tag antenna distances of 4 cm, 6 cm and 8 cm from the

reader antenna.

3D pattern – To perform the experiment two acrylic plates are used. The reader

antenna is fixed in the reader plate using tape. A tag antenna plate is made by making

a 26 cm x 26 cm square pattern of holes spaced 1 cm apart. The tag antenna is moved

along these holes at distances of 2 cm apart and the efficiency of power transfer is

measured. At the end the efficiency pattern is obtained for 13 cm x 13 cm square

matrix. The reader antenna is fixed and the tag antenna plate is moved at distances of

4 cm, 6 cm and 8 cm from the reader antenna plate and the efficiency pattern is

obtained. For references and efficiency pattern plotting, two axis – the x axis and y axis

are used. The vertical upward and downward movement of the tag antenna in the tag

antenna plate is considered negative and positive movement of the tag antenna along

the x axis. The horizontal east and west movement of the tag antenna is considered

positive and negative movement of the tag antenna along the y axis.

4.3 Antenna characterization setup requirements

In order to obtain an accurate characterization [12] of an antenna, the following steps need

to be taken.

The system needs to be very stable. This can be realized by building a system that requires

the least human intervention. The setup should be such that, the reliability of the measurements

should be maintained. Hence the experiments performed should be repeatable. The setup

38

should be small and easy to handle to promote ease of experiments with minimum errors. This

would require the setup to be reconfigurable. The setup should be such that, a large number of

readings from the setup can be taken in a short interval of time. Unnecessary hardware

components should be removed to reduce hardware costs. A better measurement sensitivity

capability should be maintained in order to obtain experimental accuracy.

4.4 Experimental setup design

In order to realize such a system, where the above conditions are met, a design was made

and was accurately fabricated using a CNC machine. The design of the setup for one groove is

shown in Figure 4.2.

Figure 4.2 Top view of entire setup with a schematic of rotation angle of reader antenna

In order to make the setup reconfigurable and, as per the dimensions mentioned in Figure 4.3,

three grooves were made of diameters equal to the diameters of 3 pvc pipes. Three holes were

made at distances 4 cm, 6 cm and 8 cm from the 0 degree position (origin) mentioned in Figure

4.2. These holes were made to fit the acrylic stick into which tag antenna would be fixed using

tooth pick and tape. A schematic of the three grooves and holes is shown in Figure 4.3.

39

Figure 4.3 Reconfigurable setup schematic

The pvc pipe is cut into half for ease in performing experiments. The reader antenna is stuck

behind the pvc pipe which is made to fit into the respective groove and the experiments are

performed. Antenna misalignment, antenna pattern and cross section pattern are then obtained

as shown in Step 1, Step2 and Step3 of Figure 4.4.

Figure 4.4 Antenna misalignment, Antenna pattern, Cross section pattern using steps – 1, 2, 3

The 3D pattern is obtained using the setup design shown in Figure 4.8.

40

Figure 4.5 3D pattern setup design

(a) (b)

(c) (d)

(e) (f)

Figure 4.6 (a) Top view, (b) Side view of setup, (c) Back view of pvc where reader antenna is

attached, (d) Top view of 3D setup, (e) Top view of 3D setup, (f) Front view of tag antenna plate

41

4.5 Experiment parameters

After achieving a good experimental set up, experiments were conducted and the

parameters used are discussed. Experiments were conducted using three curvatures C1 of radii

= 8.4 cm, C2 of radii = 11 cm and C3 of radii = 14 cm. As discussed before, three half sections

of pvc pipes were used to emulate the bend. Based on the radii C1 had the highest bend, C2

had the second highest bend, and C3 had the least bend. Antennas with two different designs

were used in the experiments. Since with bending of the coil, antenna inductance reduces and

the transmitter circuit had to be retuned for each bend [7, 8]. The receiver circuit is changed

only once when tuning the antenna for the first time when it is straight. There after only the

transmitter circuit is retuned. Also the two antennas are abbreviated as A1 and A2. The

component parameters are mentioned in Table 4.1.

42

Table 4.1 Experiment parameters

4.6 Antenna characterization results

Legends shown in the graphs of Figure 4.7 – Figure 4.26 and Figure 4.30 – Figure 4.49,

consists of 4 parts separated by commas. The first part is the antenna used – A1 or A2. The

second part is the curvature used – C1, C2, C3 or St (straight). The third part is the distance

between the reader coil and tag coil. The fourth part is the experiment conducted – A.M

(Antenna Misalignment), A.P (Antenna Pattern), C.P (Cross-section Pattern) and 3D.P (3D

Pattern).

43

Results obtained using Antenna 1 are shown in Figure 4.7 – Figure 4.29.

Figure 4.7 Antenna pattern distance comparison for A1 with flat bend

Figure 4.8 Cross-section pattern distance comparison for A1 with flat bend

The maximum efficiency obtained using the straight reader antenna is around 12 - 13 % when

the two coils are center aligned. From the comparison graphs, it is observed that the efficiency

reduces with distance. The efficiency pattern has a gaussian shape for misalignment pattern,

antenna pattern and cross section pattern experiments. The graphical results for the least bend

curvature i.e. C3 are shown in Figure 4.9 – Figure 4.11.

44

Figure 4.9 Antenna misalignment distance comparison for A1 with C3 bend

Figure 4.10 Antenna pattern distance comparison for A1 with C3 bend

45

Figure 4.11 Cross-section pattern distance comparison for A1 with C3 bend

The maximum efficiency obtained with the reader antenna bent at C3 with the two coils center

aligned increases to around 14 %. From the comparison graphs, it is observed that the

efficiency reduces with distance. The efficiency pattern has a gaussian shape for misalignment

pattern, antenna pattern and cross section pattern experiments. The graphical results for the C2

bend are shown in Figure 4.12 – Figure 4.14.


46



The maximum efficiency obtained with the reader antenna bent at C2 with the two coils center

aligned increases to around 16 % - 17 %. From the comparison graphs, it is observed that the

efficiency reduces with distance which can be explained from inverse square law. The

efficiency pattern has a gaussian shape for misalignment pattern, antenna pattern and cross

47

section pattern experiments. The graphical results for C1 bend is shown in Figure 4.15 – Figure

4.17.



48


The maximum efficiency obtained with the reader antenna bent at C1 with the two coils

center aligned is around 18 %. From the comparison graphs, it is observed that the efficiency

reduces with distance. The efficiency pattern has a gaussian shape for misalignment pattern,

antenna pattern and cross section pattern experiments. Graphs of different curvatures are then

compared keeping the parameter, distance as constant.

Figure 4.18 Antenna misalignment curvature comparison for A1 at 4cm distance

49

Figure 4.19 Antenna misalignment curvature comparison for A1 at 6 cm distance


Figure 4.21 Antenna pattern curvature comparison for A1 at 4 cm distance

50



Figure 4.24 Cross-section pattern curvature comparison for A1 at 4 cm distance

51



From the curvature comparison results shown in Figure 4.18 – Figure 4.26, it is observed that

the bandwidth increases with increase in bend and increase in distance. 3D plotting

experiments were conducted later to obtain a plot at distances of 4 cm, 6 cm and 8 cm. 169

data points were collected at each distance.

52

Figure 4.27 3D pattern for A1 at 4 cm distance


53


The above graphs show that a gaussian shaped curve is obtained at the efficiency

measurements for each column. The gaussian shaped curve for the central column has the

highest magnitude efficiency values. And the magnitude of the Gaussian curve decreases on

either side of the central column.

In order to ascertain that all the graphs obtained for antenna A1 can be repeated for

another flexible antenna design, a new antenna A2 with a different design mentioned in Table

4.1 was fabricated. The same number and type of experiments were conducted using A2. The

results are shown in Figure 4.30 – Figure 4.52.

54

Figure 4.30 Antenna pattern distance comparison for A2 with flat bend

Figure 4.31 Cross-section pattern distance comparison for A2 with flat bend

The maximum efficiency obtained using the straight reader antenna is around 13 % – 14 %

when the two coils are center aligned. From the comparison graphs, it is observed that the


pattern, antenna pattern and cross section pattern experiments. The graphical results for the

least bend curvature i.e. C3 are shown in Figure 4.32 – Figure 4.34.

55



56



center aligned increases to around 15 %. From the comparison graphs, it is observed that the


pattern, antenna pattern and cross section pattern experiments. The graphical results for C2

bend is shown in Figure 4.35 – Figure 4.37.


57




center aligned increases to around 16 % - 17 %. From the comparison graphs, it is observed

that the efficiency reduces with distance. The efficiency pattern has a gaussian shape for

misalignment pattern, antenna pattern and cross section pattern experiments. The graphical

results for C1 bend is shown in Figure 4.38 – Figure 4.40.

58




59


center aligned increases to around 18%. From the comparison graphs, it is observed that the

efficiency reduces with distance which can be explained from inverse square law. The

efficiency pattern has a gaussian shape for misalignment pattern, antenna pattern and cross

section pattern experiments. Graphs of different curvatures are then compared keeping the

parameter, distance as constant.



60




61




62


From the curvature comparison results shown in Figure 4.41 – Figure 4.49, it is observed

that the bandwidth increases with increase in bend and increase in distance. 3D plotting

experiments were conducted later to obtain a plot at distances of 4 cm, 6 cm and 8 cm. 169

data points were collected at each distance.


63



64

The above graphs show that a Gaussian curve is obtained at the efficiency measurements for

each column. The Gaussian curve for the central column has the highest magnitude efficiency

values. And the magnitude of the Gaussian curve decreases on either side of the central

column.

4.7 Explanation of the results

Faraday, showed that the magnitude of electromagnetic force generated in the coil is

proportional to the rate of change of magnetic flux passing through all the turns of the coil. With

the displacement of either antenna that causes misalignment between the centers of the two

coils, the passage of magnetic flux through all the turns of the Tag antenna reduces, thus

causing the reduction in emf generated at the tag coil, hence the reduction in efficiency. This

phenomenon is verified by the graphs obtained.

It was found from experiments that by using a straight coil, the efficiency obtained is

around 12% which is less than other curvatures used. By increasing the distance between

reader antenna and the tag antenna the efficiency decreases, because of the inverse square

law. From the graphs, it is observed that the efficiency of power transfer decreases on either

side of the zero position of the reader antenna. The curve is Gaussian shaped, and since the

power transferred between the two antennas is due to change in magnetic flux linking the two

coils, the intensity of magnetic field distribution on either side of the zero position is gaussian.

Gaussian distribution of magnetic field intensity is observed, when both the reader and tag coils

are rotated, keeping either antenna static when the other is rotated. Also the cross section

pattern follows a gaussian distribution, indicating that the magnetic flux density is strongest at

the center of the antenna and reduces in a gaussian manner, as the tag antenna is moved away

from the center. From, the comparison graphs shown above, a summarized picture of the

magnetic field distribution is obtained. The 3D pattern gives a better picture of the magnetic field

distribution between the reader and tag antennas. From the gaussian shaped curves obtained

by moving the tag across each row of the tag plate, it is understood how the magnetic field

65

varies in the three dimensional region between the reader antenna and tag antenna. The

gaussian shape of the curves can be explained by the below formulas [31].

2

( , )( , )

4t

GW P

r

(4.4)

where W is the power density of the incident radiation from the reader antenna. and

indicate the direction dependence of the parameters. G is the gain of the reader antenna which

is dependent on the radiation efficiency, antenna gain and mismatch losses between reader

antenna and transmitter circuit. tP is the power the reader antenna would deliver to a matched

load. r is the distance between the reader and tag antennas.

( , )rP A W (4.5)

where rP is the power delivered to the tag antenna, A is the effective aperture of the tag

antenna which is the effective area of the tag antenna that directly intercepts the power

transmitted by the reader antenna.

24

r

t

P GA

P r (4.6)

(4.6) is obtained from (4.4) and (4.5) and is the efficiency of power received at the tag antenna.

The efficiency depends on the effective aperture of tag antenna, gain of the reader antenna and

the distance between the tag and reader antennas. The gain of the reader antenna is constant

since the source is constant. Hence the curves obtained in Figure 4.7 – Figure 4.52 are

dependent on the effective aperture of the tag antenna and the distance between the tag and

reader antennas.

By increasing the bend of the coil by using pvc pipes, an efficiency of around 14%, 16%

and 18% is obtained between center-aligned reader and tag antennas. Thus there is

approximately, an increase of 2% in the power transfer efficiency with the increase in the

bending radius of the coil by 4 cm. This hints that the magnetic flux at the center of the two coils

66

increases due to contribution from the sides of the reader antenna. And with the increase in

bend, the contribution made from either side of the reader antenna center, increases.

67

CHAPTER 5

CONCLUSION

Simulation model was established using Ansoft HFSS program. To ensure simulation

results obtained from Ansoft HFSS software are accurate a spiral antenna simulation model had

to be established. Planar spiral antenna formulae referred from [17] were used. Since the

results obtained from these formulae are only valid at dc frequencies, different spiral antenna

designs were used and simulated to match the results obtained from the formulae. The results

matched. Two IEEE papers [16, 24] were used to ascertain whether the model is correct.

Inductance and quality factor results obtained from these two papers also matched with the

appropriate models made in HFSS.

The experimental results validated simulation model. The next step was to ascertain

whether the experimentally fabricated inductor results matched the HFSS simulation results.

Two different substrates, wood and kapton, were used and 35 experiments were conducted

using inductors of different design. The simulation results matched the HFSS model in all the

experiments.

Power transfer efficiency and radiation pattern were experimentally tested. A total of

125 experiments were conducted using a stable well designed setup. Experiments were

conducted using two different reader antennae, three different bending angles of reader

antenna at three different distances. All results obtained were compared with results obtained

from a straight reader antenna. 3D radiation pattern experiments were conducted and the

results obtained, establish the radiation pattern in the space between the reader and tag

antenna. Using four different experiments the power transfer efficiency was obtained and the

radiation pattern was thus established.

68

Flat and curve antenna with three different curvatures were tested. Three different

curvatures C1 of 8.4 cm bend, C2 of 11.4 cm bend and C3 of 14 cm bend were used in the

characterization experiments and the reader antenna was bent along these curvatures and

tested. The antenna pattern comparison results show that the pattern does not change with

bending. Also the efficiency increases with increase in bend.

The antenna misalignment experiment suggests that when the patient wears the reader

antenna in the wrong position the radiation pattern changes for different degree of reader

antenna bend. The antenna pattern experiment suggests that when the tag antenna inside the

body moves along its axis the radiation pattern does not change for different degree of reader

antenna bend. The cross-section pattern experiment suggests that when the doctor commits a

mistake in aligning the tag antenna with the reader antenna while inserting the implant inside

the patient using endoscopy, the radiation pattern does not change for different degree of

reader antenna bend. The 3D radiation pattern experiment indicates the working zone of the tag

antenna when it is attached to a sensor. From these experiments, it can be concluded that the

radiation pattern will change if the reader antenna moves but the pattern will not change if the

tag antenna moves for different degree of reader antenna bend. These results prove that

flexible antennas can be used for wireless powering of medical implants since bending the

antenna only improves the efficiency and does not change the antenna pattern.

69

CHAPTER 6

FUTURE WORK

6.1 Agilent module

In the thesis work discussed, all characterization results were obtained manually. This

involved weeks of work in collecting and processing data, the time spent in fabricating the setup

etc. A major chunk of the time was consumed in ensuring that the reader and tag antennas are

perfectly aligned at the respective orientations. This involved repetition of the results to ensure

stable graphical results. Several companies have built different types of modules involving high

– tech gadgets for accurately obtaining thousands of data points in a short period of time. One

such gadget is the one built by Agilent [12]. A pictorial diagram of the module is shown in Figure

6.1.

Figure6.1 Agilent module[12]

From the module shown, the bending angle of the flexible antenna can be controlled. The power

pattern can be obtained in the 3 dimensional space surrounding the antenna by using a scanner

controller device. The PNA – X network analyzer can be configured for both near-field and far-

70

field measurements. The data collected by the network analyzer can then be stored in a

computer.

More than forty thousand data points can be obtained in a time interval of 10 to 20 minutes,

thus reducing the experimental time by one tenths. Also there is no human intervention, since

the experiments conducted are controlled automatically, thus improving the reliability of the

results obtained. Optimum Measurement sensitivity is a critical aspect for antenna

measurements, which is provided by the device.

In the future, reliable characterization experiments can be conducted using this module

for any kind of antenna in a short interval of time.

6.2 Stacked spiral antennas

By making use of flexible stacked antennas, the inductive power transfer can be

increased and the resonant magnetic coupling frequency can be reduced [10, 30], thereby

reducing the power absorbed by the body tissues and increasing the range of operation.

Characterization of these antennas is therefore important.

Figure 6.2 Stacked spiral antenna [10]

71

APPENDIX A

MATERIAL DATA SHEETS AND SPECIFICATIONS

72

A.1 KAPTON DATA SHEET

Kapton is a biocompatible material used as a substrate for fabrication of flexible antennas. Their

characteristics are discussed in Table 1, Table 2 and Table 3.

TABLE 1 Typical Electrical Properties of DuPont™ Kapton

® HN Film at 23°C (73°F), 50% RH

TABLE 2

Thermal Properties of DuPont™ Kapton® HN Film

73

TABLE 3

Physical Properties of DuPont™ Kapton® HN at 23°C (73°F)

74

A2. JVCC CFL-5CA COPPER FOIL TAPE (CONDUCTIVE ADHESIVE) SPECIFICATIONS

Adhesive: The bottom side of copper foil has conductive acrylic with the help of which it can

stick on any material

Carrier/Backing: copper foil

Release Liner: 60# paper

Thickness of tape: 3 mils (without liner) 1.6 mils (adhesive) 1.4 mils (carrier/backing)

Adhesion: 70 ounces per inch (to stainless steel)

Tensile Strength: 36 pounds per inch (longitudinal)

Service/Operating Temperature: up to 311°F

Elongation: 6%

Electrical Resistance: 0.003 Ohms/sq. in.

Certifications: File #130121, UL510, Section 4, MIL-T-47012

Core: 3" diameter neutral

http://www.findtape.com/product329/JVCC-CFL-5CA-Copper-Foil-Tape.aspx?tid=3&info=3M+Scotch+SJ5300+Series+Bumpon+Self-Adhesive+Bumpers

75

APPENDIX B

MATLAB PROGRAMS

76

B1. CURRENTSHEET AND WHEELER FORMULAE

Matlab [14] codes were written for speedy calculation of inductances using formulas mentioned

in [17].

Currentsheet Function

function [Lgmd]=currentsheet(n,din,dout)

mu = 4*3.141592654*(10^-7);

c1 = 1.27;

c2 = 2.07;

c3 = 0.18;

c4 = 0.13;

davg = (dout+din)*0.5*(10^-3);

p = (dout - din)/(dout + din);

Lgmd = ((0.5*mu*(n^2)*davg*c1)*(loge(c2/p)+(c3*p)+(c4*(p^2))))*(10^6);

End

Wheeler Function

function [Lmw]=wheeler(n,din,dout)

k1 = 2.34;

k2 = 2.75;

davg = 0.5*(10^-3)*(dout+din);

p = (dout - din)/(dout + din);

Lmw = (k1*4*3.141592654*(10^-7)*(((n^2)*davg)/(1+(k2*p))))*(10^6);

End

77

B2. 3D PATTERN MATLAB PROGRAM

Also a matlab code was written to plot the 3D pattern graphs. The value of X axis and Y axis

was given as input. The value of Z axis which is the efficiency values at corresponding value of

X and Y are also given as input. The 3D pattern graph was then obtained using the in-built

matlab function -

plot3 (X,Y,Z)

legend (‘Ax, d = x cm’).

78

REFERENCES

[1] Upkar Varshney “Pervasive Healthcare and Wireless Health Monitoring”, Journal of Mobile

Networks and Applications, Volume 12, Issue 2-3, March 2007, Pages 113 – 127.

[2] Xiaoyu Liu, Fei Zhang, Steven A. Hackworth, Robert J. Sclabassi and Mingui Sun “Wireless

Power Transfer System Design for Implanted and Wom Devices”, Bioengineering Conference,

2009 IEEE 35th Annual North East.

[3] Anthony N. Laskovski, Tharaka Dissanayake and Mehmet R. Yuce “Wireless Power

Technology for Biomedical Implants”, Biomedical Engineering, pp 119 – 132.

[4] James W. (“Rus”) Healy “Antenna Here is a Dipole”, NJ2L.

[5] Thermpon Ativanichayaphong “Wireless Devices for Medical Applications”, ProQuest, The

University of Texas at Arlington, Electrical Engineering, December 2007.

[6] Anders J Johansson “Wireless Communication with Medical Implants : Antennas and

Propagation”, Department of Electroscience, Lund University, June 2004.

[7] Jaime Garnica, Joaquin Casanova, and Jenshan Lin “High Efficiency Midrange Wireless

Power Transfer System”, IMWS-IWPT2011 Proceedings.

[8] Teck Chuan Beh, Takehiro Imura, Masaki Kato, Yoichi Hori, “Wireless Power Transfer

System via Magnetic Resonant Coupling at Restricted Frequency Range – Fixing Resonace

Frequency With Impedance Matching”, University of Tokyo, 2010.

[9] Nathan O. Sokal “Class – E RF Power Amplifiers”, WA1HQC of Design Automation, Inc,

Jan/Feb 2001.

[10] Anthony Nikola Laskovski, Mehmet Rasit Yuce, and Tharaka Dissanayake, “Stacked

Spirals for Biosensor Telemetry”, IEEE Sensors Journal, VOL. 11, No. 6, June 2011.

79

[11] P. R. Troyk and G. A. DeMichele, “Inductively-coupled power and data link for neural

prostheses using a class-E oscillator and FSK modulation,” IEEE International Conference

Engineering in Medicine and Biology Society, Vol.4, pp.3376-3379, 2003.

[12] Agilent Technologies “Solutions for Antenna Characterization”

[13] Ansoft Corporation “Ansoft High Frequency Structure Simulator v10 User’s Guide”, 2005.

[14] Christos Xenophontos “A Beginner’s Guide to MATLAB”, Department of Mathematical

Sciences, Loyola College.

[15] Oleksiy Klymenko “UWB Receiver Frontend Design with ADS”, ADS User Group Meeting,

IHP, Frankfurt, Germany, pp 13 – 15, May 14 & 15, 2009.

[16] Uei – Ming Jow, Maysam Ghovanloo “Modeling and Optimization of Printed Spiral Coils in

Air, Saline, and Muscle Tissue Environments”, IEEE Transactions on Biomedical Circuits and

Systems, Vol . 3, No. 5, October 2009.

[17] Sunderajan S.Mohan, Maria del Mar Hershenson, Stephen P. Boyd, and Thomas H.Lee

"Simple Accurate Expressions for Planar Spiral Inductances", IEEE Journal of Solid-State

Circuits, Vol. 34, No. 10, October 1999.

[18] AutoDesk Official Training Guide “Autocad 2010 Volume 1”, April 2009.

[19] AutoDesk Official Training Guide “Autocad 2010 Volume 2”, April 2009.

[20] AutoDesSys “Bonzai 3d Online Manual”.

[21] http: www.antenna-theory.com.

[22] Constantine A. Balanis "Antenna Theory: Analysis and Design", 3rd Edition, 1999.

[23] Enver G. Kilinc, Catherine Dehollain, "Design and Optimization of Inductive Power

Transmission for Implantable Sensor System", Symbolic and Numerical Methods, Modeling and

Applications to Circuit Design (SM2ACD), 2010 XIth International Workshop on 4-6 Oct, 2010.

[24] Eui-Jung Yun, Jae-Wook Kim, Won-Gook Lee, Young-Jin Kim, Hyeong-Sik Park "Analytical

estimation of high-frequency properties of RF micro-inductors prepared by direct-write

techniques", Springer Science + Business Media, LLC 2007.

http://www.antenna-theory.com/

80

[25] M. Bartek, S.M. Sinaga, J.N. Burghartz "Influence of via-connections on electrical

performance of vertically-spaced RF passives", Electronic Components and Technology

Conference, 2005.

[26] Amal Zaki "Design and Fabrication of High-Q Spiral Inductors Using MEMS Technology",

Proceedings of the 4th WSEAS Int. Conference on Electromagnetics, Wireless and Optical

Communications, Venice, Italy, November 20-22, 2006.

[27] Jr-Wei Lin, C. C. Chen, and Yu-Ting Cheng, Member, IEEE "A Robust High-Q

Micromachined RF Inductor for RFIC Applications", IEEE Transactions on Electron Devices,

Vol. 52, No. 7, July 2005.

[28] Kenichi Okada and Kazuya Masu "Modeling of Spiral Inductors", Tokyo Institute of

Technology, Japan.

[29] Steve McFee and Dennis Giannacopoulos, "Introduction to Adaptive Finite Element

Analysis for Electromagnetic Simulations", Technical Article, Department of Electrical &

Computer Engineering, McGill University, Canada.

[30] Anthony Nikola Laskovski, Mehmet Rasit Yuce, Tharaka Dissanayake "Stacked Spirals for

Biosensor Telemetry", IEEE Sensors Journal, Vol. 11, NO. 6, June 2011.

[31] http:www.google.com

81

BIOGRAPHICAL INFORMATION

Safwan Ahmed Muneer was born in Bangalore, Karnataka, India. He obtained his

Bachelor’s degree in Electronics and Communication Engineering in 2009, from Visweswaraiah

Technological University, Belgaum, Karnataka, India. He has worked towards his Master of

Science in Electrical Engineering in The University of Texas at Arlington since August 2010. His

Research areas include Antenna Design and Wireless radio frequency Communication.

Wireless Power for Implasnts

Documents

thesis work

body implant powering

research work

new kind of body implants

retinal implants

university of texas

thesis committee members

support of